i
Evaluating the design of tax-free investments to
promote household savings in South Africa: an
international comparison
MH Groenewald
orcid.org 0000-0002-0917-4539
Mini-dissertation submitted in partial fulfilment of the
requirements for the degree
Masters of Commerce in South
African and International Taxation
at the North West University
Supervisor: Mnr H van Dyk
Graduation May 2018
ACKNOWLEDGEMENTS
I would, first, like to thank the Lord for guiding me and giving me the strength to complete this research.
I would furthermore like to express my gratitude towards the following persons:
My supervisor, Mr Herman van Dyk, for his guidance, immense knowledge and valuable assistance.
My husband, Hennie, for his love, encouragement and support. My little boy and girl, Henko and Milané, for their unconditional love.
My parents, Koos and Melanie Nortjé, who has always provided me with so many opportunities, love and support.
Elmarie Viljoen-Massyn, for assisting me with the language editing.
REMARKS
The reader is reminded of the following:
This dissertation is presented in article format, in accordance with the policies of the North-West University’s Faculty of Economic and Management Sciences, and consists of one research article. The research article and dissertation comply with the style requirements of the American Psychological Association (APA).
KEYWORDS
ABSTRACT
South African household savings have declined over the past 50 years and are currently at negative levels. The South African government has taken steps to ameliorate this problem by offering an incentive through the income tax system by way of tax-preferred savings accounts. The purpose of this study is to evaluate the design of this tax incentive in its ability to encourage South African taxpayers to increase household savings. This study is conducted in terms of the normative legal science discipline and identifies attributes or criteria which could serve as benchmarks for evaluating this new tax incentive to promote household savings. The study further evaluates the incentive against the design of an established comparable incentive from Canada. The evaluation found that, although the incentive has some mechanisms in place to encourage new savings, only a fifth of accounts opened during the first year of operation are estimated to be from new savers. While it does not disparagingly benefit wealthy individuals, the incentive does not contain any features to ensure use by low to moderate taxable income households (the target group). This study estimates that, owing to the design of the incentive where contributions are limited, the initial cost of the incentive to the fiscus will be significantly lower than the current interest-free exemption, and the real cost of the new incentive will only be borne by the fiscus in the future. It was also found that the new incentive is designed to be visible, simple, transparent and suitable. It is more equitable than its predecessor because it allows for a wider range of investments to be made tax-free. The incentive does not infringe on the certainty and convenience principles, but it remains unclear whether the incentive will be economically efficient and cost-effective. A comparison with the Canadian incentive revealed that it could be beneficial to allow for unused contributions to be carried forward, as this may lead to increased future savings. In addition, allowing replacement of withdrawals will likely enhance short-term savings, as the current incentive is more suitable for long-short-term investment.
TABLE OF CONTENTS ACKNOWLEDGEMENTS ... ii REMARKS .. ... iii KEYWORDS ... iv ABSTRACT . ... v TABLE OF CONTENTS ... vi CHAPTER 1: INTRODUCTION ... 1 1.1 INTRODUCTION ... 1
1.2 RESEARCH QUESTION AND OBJECTIVES ... 3
1.3 MOTIVATION FOR SELECTING CANADA FOR THE INTERNATIONAL COMPARISON 3 1.4 RESEARCH METHOD ... 4
1.5 OVERVIEW OF CHAPTERS ... 5
CHAPTER 2: RESEARCH ARTICLE ... 6
ABSTRACT . ... 7
KEYWORDS ... 8
2.1 INTRODUCTION ... 9
2.2 RESEARCH QUESTION AND OBJECTIVES ... 11
2.3 MOTIVATION FOR SELECTING CANADA FOR THE INTERNATIONAL COMPARISON …… ... 11
2.4 RESEARCH METHOD ... 12
2.5 DESIGN CRITERIA FOR AN EFFECTIVE SAVINGS TAX INCENTIVE ... 13
2.6 EVALUATION OF THE DESIGN OF SECTION 12T OF THE INCOME TAX ACT ... 14
2.7 COMPARISON TO THE INCENTIVE AVAILABLE IN CANADA ... 20
2.8 CONCLUSION AND RECOMMENDATIONS FOR FURTHER RESEARCH ... 23
REFERENCES ... 25
CHAPTER 3: CONCLUSION ... 29
3.1 OBJECTIVE OF THIS CHAPTER ... 29
3.2 RESEARCH FINDINGS ... 29
3.2.1 Research objective 1 ... 29
3.2.2 Research objective 2 ... 29
3.2.3 Research objective 3 ... 30
3.3 OVERALL CONCLUSION AND RECOMMENDATIONS FOR FURTHER RESEARCH .. 30
1
CHAPTER 1: INTRODUCTION
1.1 INTRODUCTION
Decline in South African household savings
South African net household savings1 has considerably declined over the past half century and is
currently at negative levels. The World Bank (2011) observed that net household savings in South Africa is critically low when compared with other developing countries. Similarly, a 2014 survey revealed that only 20% of South African adults have some form of savings with a formal financial institution (Finmark Trust, 2014).
According to National Treasury (2012), increased household savings will diminish the financial vulnerability of households and reduce reliance on debt, whereas low savings rates will result in increased reliance on foreign funds to finance investment (Aron & Muellbauer, 2000). According to Prinsloo (2000), adequate savings is vital in sustaining a “macroeconomic balance and the maintenance of financial and price stability” (p. 2). An improved savings rate is viewed as a prerequisite for sustainable economic growth (National Treasury, 2014).
Section 27(1)(c) of the Constitution (1996) guarantees all citizens the right to access to “social security, including, if they are unable to support themselves and their dependants, appropriate social assistance” (p. 14). Accordingly, the South African government has prioritised spending on social security (National Treasury, 2017). Government spending on social grants currently amounts to R151.6 billion, which represents 9.7% of consolidated government expenditure (National Treasury, 2017). Expenditure on social grants is expected to increase to R175.6 billion in the 2020 fiscal year when an estimated 18.1 million beneficiaries will receive such grants (National Treasury, 2017).
Cronje (2017) reports that South Africa is currently facing a budget shortfall of R50 billion, together with increased pressure on government to provide free tertiary education, which would place, in addition to the current allocation of R25 billion, a further burden of R71 billion on the fiscus (Phungo, 2015). The significant budget deficit, coupled with ever-increasing expenditure demands, will likely hamper government’s ability to continue delivering the constitutionally mandated social assistance to those citizens who are unable to support themselves and their dependants.
1 The Organisation for Economic Co-operation and Development (OECD, 2017) defines “net household savings” as “the subtraction of
household consumption expenditure from household disposable income, plus the change in net equity of households in pension funds” (Household savings, para. 1).
Tax incentive to promote household savings
Extant South African tax legislation contains several provisions aimed at promoting certain taxpayer behaviour. Current tax incentives include favourable tax treatment of research and development expenditure2, payroll tax credits for employing young persons3, accelerated capital allowances for
investing in new movable manufacturing assets4, preferential tax rates for certain companies
operating in special economic zones5 and zero-rating of value-added tax on certain exported goods
and services6.
The use of tax incentives to drive desired behaviour is contentious. According to Easson and Zolt (2002), the conventional view on tax incentives is that they are “bad in theory because they distort investment decisions” and “bad in practice because they are often ineffective, inefficient and prone to abuse” (p. 1). Tax incentives result in revenue losses both directly (revenue foregone to the fiscus as a result of the particular incentive) and indirectly (as a result of administrative resources being diverted from revenue collection) (Bolnick, 2004). Such revenue losses could result in dire fiscal adjustments in the form of increased taxes on other taxpayers and reductions in expenditure (Bolnick, 2004). Furman (2008) notes that a neutral tax system should not distort taxpayer behaviour. Even the South African government concedes that “well-designed incentives may not be an effective way to address underlying problems that are not tax related” (National Treasury, 2017, p.127). However, Easson and Zolt (2002) contend that all taxes result in some level of distortion and that tax incentives may be useful in correcting market failures. Applying this view to the savings predicament in South Africa, one could argue that some taxpayers might be reluctant to save because the return on their savings will be eroded by the imposition of taxation. The introduction of a savings tax incentive could alleviate their reluctance.
A new tax incentive to promote savings by South African individuals was introduced, with effect from 1 March 2015, in terms of section 12T of the Income Tax Act (58 of 1962). Similar to the Individual Savings Accounts (ISAs) incentive implemented in the United Kingdom, the South African tax incentive allows taxpayers to make limited contributions to tax-preferred savings vehicles, of which all returns are tax-free (National Treasury, 2014). Contributions are made from post-tax income and may be withdrawn at any time, but replacement of withdrawals depletes the contribution limit. The ISAs incentive has attracted a substantial number of not only low- to moderate-income investors in
2 Section 11D of the Income Tax Act (58 of 1962). 3 Employment Tax Incentive Act (26 of 2013). 4 Section 12C of the Income Tax Act (58 of 1962). 5 Section 12R of the Income Tax Act (58 of 1962). 6 Section 11 of the Value-added Tax Act (89 of 1991).
the UK, but also high-income earners (National Treasury, 2012). Other jurisdictions that have introduced similar incentives include the United States of America and Canada.
Both National Treasury (2014) and Easson and Zolt (2002) emphasise the importance of the appropriateness of the design of incentives to achieve their intended objective, especially considering recent revelations from behavioural economics.
1.2 RESEARCH QUESTION AND OBJECTIVES
The research question that this study aims to address is the following: Has the tax incentive in section 12T of the South African Income Tax Act been designed appropriately to promote household savings amongst South African individual taxpayers?
The following research objectives have been formulated to address this research question:
To determine appropriate design attributes or criteria to serve as benchmarks for the evaluation of an effective savings tax incentive;
To evaluate the design of section 12T against such attributes or criteria to determine whether the design of the statute is fundamentally effective to promote household savings in South Africa; and
To evaluate the design of section 12T against the comparable incentive already introduced in Canada.
1.3 MOTIVATION FOR SELECTING CANADA FOR THE INTERNATIONAL COMPARISON
Canada has a similar tax incentive to encourage savings called the “tax-free saving account” (TFSA). Earnings from a TFSA is free from tax, and contributions also have certain set contribution limits (Alarie, 2009) similar to the tax-free investment available in South Africa.
According to Scarlat and Lefebvre (2008), the OECD identifies a high domestic savings rate as an important tax policy. Canada introduced the TFSA to align itself with other OECD countries which use tax incentives to increase domestic savings (Scarlat & Lefebvre, 2008).
Canada is considered to be a developed country, while South Africa is considered to be a developing country (UN Statistics Division, n.d.). Canada is a high-income country, while South Africa is an upper middle-income country (World Bank, 2016). A study of the design of the tax incentive of Canada will therefore contribute to improving the tax incentive available in South Africa.
Approximately 4.9 million individuals have invested in TFSAs by the end of 2009, 6.7 million by the end of 2010 (Kesselman, 2015), and 8.2 million by the end of 2011 (Department of Finance Canada, 2012). Individuals over all income levels have invested in TFSAs, while individuals with income less than $80 000 made up approximately 80% of TSFA holders and contributions during 2011 (Department of Finance Canada, 2012). In 2013, almost 11 million Canadians have invested in a TFSA (Kesselman, 2015). Canada’s TFSA has therefore proven to encourage savings, as more individuals have been saving each year since its introduction.
Canada was therefore selected for the comparative study, the legislation provisions of the tax incentive in Canada is very similar to that of South Africa, and therefore its TFSA’s design will be compared with the tax-free investment in South Africa. A study of the design will contribute to improving the tax incentive available in South Africa with the ultimate aim to promote economic growth in this country.
1.4 RESEARCH METHOD
McKerchar (2009) describes legal research as a possible research paradigm in the field of taxation. Normative legal science, a discipline within the legal research paradigm, is focused on striving for “better law” (Van Hoecke, 2011) by asking the question “how ought the law to read?” (Smits, 2012, p. 9). In the normative approach, the law ought to comply with norms and standards that maximise welfare (Hage, 2011). Smits (2012) advocates a normative perspective external from the legislature and courts.
The purpose of this study is to evaluate the design of tax-free investments to promote household savings in South Africa. Therefore, the study is conducted in terms of the normative legal science discipline.
Consequently, the study aims to determine appropriate design attributes or criteria to serve as a norm for conducting the design evaluation. The study considers extant research on tax incentives, objectives proclaimed by government, and principles of sound tax policy to determine the appropriate norms to be used in the design evaluation.
1.5 OVERVIEW OF CHAPTERS
Chapter 1: Introduction and background, research question, research objectives and research method
This introductory chapter of the study serves to illustrate the background to and relevance of the study. This chapter states the research question and research objectives and explains the research method adopted in the study. Furthermore, it provides a motivation on why the comparison between South African and Canada is relevant within the context of the study.
Chapter 2 (research article): Evaluating the design of tax-free investments to promote household savings in South Africa: an international comparison
This chapter (the research article) evaluates the design of the new tax incentive to encourage savings against specific criteria and against a comparative Canadian incentive.
Chapter 3: Conclusion
This chapter summarises the findings of the study and addresses the research question formulated in chapter 1. Recommendations, based on the findings, are also made and suggestion for further research are highlighted.
CHAPTER 2: RESEARCH ARTICLE Leanie Groenewald
North-West University, South Africa
PostNet Unit 342, Private Bag X1288, Potchefstroom, 2520, South Africa 20099010@nwu.ac.za
Herman van Dyk
North-West University, South Africa
PO Box 19771, Noordbrug, 2522, South Africa herman.vandyk@nwu.ac.za
AUTHOR INFORMATION
Leanie Groenewald is a Senior Lecturer in Taxation at the Potchefstroom Campus of the
North-West University in South Africa. She is a qualified Chartered Accountant with the South African Institute of Chartered Accountants (SAICA) and holds a Master of Commerce in Forensic Accountancy.
Herman van Dyk is the Subject Chair for Taxation at the Potchefstroom Campus of the North-West
University in South Africa. He is a qualified Chartered Accountant with the South African Institute of Chartered Accountants (SAICA) and the Institute of Chartered Accountants in England and Wales (ICAEW). He holds a Master of Commerce in South African and International Taxation.
ABSTRACT
South African household savings have declined over the past 50 years and are currently at negative levels. The South African government has taken steps to ameliorate this problem by offering an incentive through the income tax system by way of tax-preferred savings accounts. The purpose of this study is to evaluate the design of this tax incentive in its ability to encourage South African taxpayers to increase household savings. This study is conducted in terms of the normative legal science discipline and identifies attributes or criteria which could serve as benchmarks for evaluating this new tax incentive to promote household savings. The study further evaluates the incentive against the design of an established comparable incentive from Canada. The evaluation found that, although the incentive has some mechanisms in place to encourage new savings, only a fifth of accounts opened during the first year of operation are estimated to be from new savers. While it does not disparagingly benefit wealthy individuals, the incentive does not contain any features to ensure use by low to moderate taxable income households (the target group). This study estimates that, owing to the design of the incentive where contributions are limited, the initial cost of the incentive to the fiscus will be significantly lower than the current interest-free exemption, and the real cost of the new incentive will only be borne by the fiscus in the future. It was also found that the new incentive is designed to be visible, simple, transparent and suitable. It is more equitable than its predecessor because it allows for a wider range of investments to be made tax-free. The incentive does not infringe on the certainty and convenience principles, but it remains unclear whether the incentive will be economically efficient and cost-effective. A comparison with the Canadian incentive revealed that it could be beneficial to allow for unused contributions to be carried forward, as this may lead to increased future savings. In addition, allowing replacement of withdrawals will likely enhance short-term savings, as the current incentive is more suitable for long-short-term investment.
KEYWORDS
2.1 INTRODUCTION
Decline in South African household savings
South African net household savings7 has considerably declined over the past half century and is
currently at negative levels (as reflected in Figure 1). The World Bank (2011) observed that net household savings in South Africa is critically low when compared with other developing countries. Similarly, a 2014 survey revealed that only 20% of South African adults have some form of savings with a formal financial institution (Finmark Trust, 2014).
According to National Treasury (2012), increased household savings will diminish the financial vulnerability of households and will reduce reliance on debt, whereas low savings rates will result in increased reliance on foreign funds to finance investment (Aron & Muellbauer, 2000). According to Prinsloo (2000), adequate savings is vital in sustaining a “macroeconomic balance and the maintenance of financial and price stability” (p. 2). An improved savings rate is viewed as a prerequisite for sustainable economic growth (National Treasury, 2014).
Section 27(1)(c) of the Constitution (1996) guarantees all citizens the right to access to “social security, including, if they are unable to support themselves and their dependants, appropriate social assistance” (p. 14). Accordingly, the South African government has prioritised spending on social security (National Treasury, 2017). Government spending on social grants currently amounts to R151.6 billion which represents 9.7% of consolidated government expenditure (National Treasury, 2017). Expenditure on social grants is expected to increase to R175.6 billion in the 2020 fiscal year when an estimated 18.1 million beneficiaries will receive such grants (National Treasury, 2017). Cronje (2017) reports that South Africa is currently facing a budget shortfall of R50 billion, together with increased pressure on government to provide free tertiary education, which would place, in addition the current allocation of R25 billion, a further burden of R71 billion on the fiscus (Phungo, 2015). The significant budget deficit, coupled with ever-increasing expenditure demands, will likely hamper government’s ability to continue delivering the constitutionally mandated social assistance to those citizens who are unable to support themselves and their dependants.
7 The Organisation for Economic Co-operation and Development (OECD, 2017) defines “net household savings” as “the subtraction of
household consumption expenditure from household disposable income, plus the change in net equity of households in pension funds” (Household savings, para. 1).
Figure 1: Household savings as percentage of gross domestic product (GDP)
Source: data obtained from Prinsloo (2000), National Treasury (2012) and Trading Economics (2017)
Tax incentive to promote household savings
Extant South African tax legislation contains several provisions aimed at promoting certain taxpayer behaviour. Current tax incentives include favourable tax treatment of research and development expenditure8, payroll tax credits for employing young persons9, accelerated capital allowances for
investing in new movable manufacturing assets10, preferential tax rates for certain companies
operating in special economic zones11 and zero-rating of value-added tax on certain exported goods
and services12.
The use of tax incentives to drive desired behaviour is contentious. According to Easson and Zolt (2002), the conventional view on tax incentives is that they are “bad in theory because they distort investment decisions” and “bad in practice because they are often ineffective, inefficient and prone to abuse” (p. 1). Tax incentives result in revenue losses both directly (revenue foregone to the fiscus as a result of the particular incentive) and indirectly (as a result of administrative resources being diverted from revenue collection) (Bolnick, 2004). Such revenue losses could result in dire fiscal adjustments in the form of increased taxes on other taxpayers and reductions in expenditure (Bolnick, 2004). Furman (2008) notes that a neutral tax system should not distort taxpayer behaviour. Even the South African government concedes that “well-designed incentives may not be an effective way to address underlying problems that are not tax related” (National Treasury, 2017, p.127). However, Easson and Zolt (2002) contend that all taxes result in some level of distortion and that tax incentives may be useful in correcting market failures. Applying this view to the savings predicament in South Africa, one could argue that some taxpayers might be reluctant to save because the return on their
8 Section 11D of the Income Tax Act (58 of 1962). 9 Employment Tax Incentive Act (26 of 2013). 10 Section 12C of the Income Tax Act (58 of 1962). 11 Section 12R of the Income Tax Act (58 of 1962). 12 Section 11 of the Value-added Tax Act (89 of 1991).
‐2.00% 0.00% 2.00% 4.00% 6.00% 8.00% 1960s 1970s 1980s 1990s 2000s 2010 2011 2017
Household savings as % of GDP
Savings as % of GDPsavings will be eroded by the imposition of taxation. The introduction of a savings tax incentive could alleviate their reluctance.
A new tax incentive to promote savings by South African individuals was introduced, with effect from 1 March 2015, in terms of section 12T of the Income Tax Act (58 of 1962). Similar to the Individual Savings Accounts (ISAs) incentive implemented in the United Kingdom, the South African tax incentive allows taxpayers to make limited contributions to tax-preferred savings vehicles, of which all returns are tax-free (National Treasury, 2014). Contributions are made from post-tax income and may be withdrawn at any time, but replacement of withdrawals depletes the contribution limit. The ISAs incentive has attracted a substantial number of not only low- to moderate-income investors in the UK, but also high-income earners (National Treasury, 2012). Other jurisdictions that have introduced similar incentives include the United States of America and Canada.
Both National Treasury (2014) and Easson and Zolt (2002) emphasise the importance of the appropriateness of the design of incentives to achieve their intended objective, especially considering recent revelations from behavioural economics.
2.2 RESEARCH QUESTION AND OBJECTIVES
The research question that this study aims to address is the following: Has the tax incentive in section 12T of the South African Income Tax Act been designed appropriately to promote household savings amongst South African individual taxpayers?
The following research objectives have been formulated to address this research question:
To determine appropriate design attributes or criteria to serve as benchmarks for the evaluation of an effective savings tax incentive;
To evaluate the design of section 12T against such attributes or criteria to determine whether the design of the statute is fundamentally effective to promote household savings in South Africa; and
To evaluate the design of section 12T against the comparable incentive already introduced in Canada.
2.3 MOTIVATION FOR SELECTING CANADA FOR THE INTERNATIONAL COMPARISON
Canada has a similar tax incentive to encourage savings called the “tax-free saving account” (TFSA). Earnings from a TFSA is free from tax, and contributions also have certain set contribution limits (Alarie, 2009) similar to the tax-free investment available in South Africa.
According to Scarlat and Lefebvre (2008), the OECD identifies a high domestic savings rate as an important tax policy. Canada introduced the TFSA to align itself with other OECD countries which use tax incentives to increase domestic savings (Scarlat & Lefebvre, 2008).
Canada is considered to be a developed country, while South Africa is considered to be a developing country (UN Statistics Division, n.d.). Canada is a high-income country, while South Africa is an upper middle-income country (World Bank, 2016). A study of the design of the tax incentive of Canada will therefore contribute to improving the tax incentive available in South Africa.
Approximately 4.9 million individuals have invested in TFSAs by the end of 2009, 6.7 million by the end of 2010 (Kesselman, 2015), and 8.2 million by the end of 2011 (Department of Finance Canada, 2012). Individuals over all income levels have invested in TFSAs, while individuals with income less than $80 000 made up approximately 80% of TSFA holders and contributions during 2011 (Department of Finance Canada, 2012). In 2013, almost 11 million Canadians have invested in a TFSA (Kesselman, 2015). Canada’s TFSA has therefore proven to encourage savings, as more individuals have been saving each year since its introduction.
Canada was therefore selected for the comparative study, the legislation provisions of the tax incentive in Canada is very similar to that of South Africa, and therefore its TFSA’s design will be compared with the tax-free investment in South Africa. A study of the design will contribute to improving the tax incentive available in South Africa with the ultimate aim to promote economic growth in this country.
2.4 RESEARCH METHOD
McKerchar (2009) describes legal research as a possible research paradigm in the field of taxation. Normative legal science, a discipline within the legal research paradigm, is focused on striving for “better law” (Van Hoecke, 2011) by asking the question “how ought the law to read?” (Smits, 2012, p. 9). In the normative approach, the law ought to comply with norms and standards that maximise welfare (Hage, 2011). Smits (2012) advocates a normative perspective external from the legislature and courts.
The purpose of this study is to evaluate the design of tax-free investments to promote household savings in South Africa. Therefore, the study is conducted in terms of the normative legal science discipline.
Consequently, the study aims to determine appropriate design attributes or criteria to serve as a norm for conducting the design evaluation. The study considers extant research on tax incentives,
objectives proclaimed by government, and principles of sound tax policy to determine the appropriate norms to be used in the design evaluation.
2.5 DESIGN CRITERIA FOR AN EFFECTIVE SAVINGS TAX INCENTIVE
The effectiveness of a tax incentive is typically taken to mean the extent to which it promotes the desired behaviour, in this case, household savings in South Africa. Bolnick (2004) emphasises that the quality of the desired behaviour is just as important. The quality of the tax incentive in the context of this study means the promotion of new household savings by households with low to moderate
levels of taxable income as intended by National Treasury (2012). Antolín, De Serres, and De la
Maisonneuve (2004) speculate that the effectiveness of tax incentives in promoting savings will also most likely differ amongst different income groups. An incentive should therefore be designed to attract new savings from low- to moderate-income taxpayers and should not excessively benefit wealthy taxpayers.
A further consideration is the cost of the incentive and whether the economic benefit derived from the incentive exceeds the cost (Bolnick, 2004). In a study of tax incentives, Attanasio, Banks, and Wakefield (2004) concluded that the cost of the incentives resulted in negligible increases in new savings at most and that using tax incentives to promote savings is indeed a very expensive way to encourage savings. Government emphasises that “the amount of new savings generated must be more than sufficient to offset the costs to government of revenue foregone” (National Treasury, 2012, p. 12). However, cost-effectiveness extends beyond the revenue foregone and includes the impact on tax administration and social equity (Bolnick, 2004).
National Treasury (2012) indicated that the preceding savings incentive, that is, the monetary interest exemption, was not visible enough. To ensure that taxpayers are aware of the new tax incentive, visibility should be an important attribute of the incentive. The incentive should therefore be labelled as “tax-free” and lend itself to active marketing and publicity.
Government has outlined three principles that products would have to meet in order to be eligible for the incentive. These principles are simplicity, transparency and suitability (National Treasury, 2014). While these principles are aimed at protecting the consumer rather than effectively promoting new household savings, they should nevertheless be considered when the design of the incentive is evaluated.
The principle of simplicity requires that savings products be simple to understand with regard to identifying the underlying assets and returns on the product. The transparency principle deals with
disclosure of costs and fees related to the product. Lastly, the principle of suitability entails that products be aimed at low- to moderate-income taxpayers and not be of a speculative nature (National Treasury, 2014).
Galper and Steuerle (1983) focus on three specific criteria that tax incentives for savings must meet. The first criterion is that tax benefits should not go to taxpayers who merely substitute one form of savings for another. Government’s initial publication on the incentive stresses the need to attract new savings (National Treasury, 2012). It is therefore vital that the incentive be designed to encourage new savings. The second criterion states that a tax incentive with a cap or limit provides little marginal incentive for a taxpayer who already saves in excess of the contribution limit. Since the incentive is aimed at attracting new savings from low to moderate earners of taxable income who likely do not have any form of current savings, the second criterion is not considered as an imperative design attribute of the South African incentive. The third criterion is the prevention of tax arbitrage. Galper and Steuerle (1983) explain that the incentive “must provide symmetrical treatment of positive saving on the one hand and negative saving or borrowing on the other” (p. 2). A taxpayer should therefore not be able to borrow funds to acquire an investment yielding tax-exempt income while being able to deduct the interest incurred on the borrowings.
In addition to the criteria specific to tax incentives, it should also be considered whether the tax incentive conforms to the principles of sound tax policy. The four maxims penned by Smith (1776/1892) are still advocated by many as the principles of sound tax policy (Alley & Bentley, 2005; Blough, 1955; Brunori, 1997; Lynn, 1976; Reese, 1980). These four maxims are equity, certainty, economic efficiency, and convenience.
In summary, the criteria to be used in this study for evaluating the design of the new savings incentive are that it must promote new savings among low- to moderate-income earners, it must be cost-effective, it must be visible, product offerings must be simple, transparent and suitable, and it must prevent tax arbitrage. In addition, the design of the incentive must conform to the principles of equity, certainty, economic efficiency, and convenience.
2.6 EVALUATION OF THE DESIGN OF SECTION 12T OF THE INCOME TAX ACT
Evaluation of the predecessor savings incentive provision
Prior to the March 2015 enactment of the new savings incentive provision, the primary savings incentive contained within the South African Income Tax Act for South African source interest was a tax-free interest income threshold in the form of an exemption of R23 800 for taxpayers under the
age of 65 and R34 500 for taxpayers aged 65 and older (section 10(1)(i)). However, this provision was not originally designed to serve as an incentive but instead as a tool to relieve the administrative inconvenience of declaring nominal amounts of local interest on individual taxpayer’s tax return (National Treasury, 2012). The exemption was only available for South African source interest and therefore discriminated against taxpayers who opted to invest in equity instruments13 or property14. A
further criticism was the incentive’s lack of visibility, as interest-bearing products were not advertised as being “tax-free”. This incentive cost the fiscus R2.1 billion during the 2014/2015 fiscal year (National Treasury, 2017), but was straightforward and easy to administer (National Treasury, 2012). The initial intention was to replace the tax-free interest income threshold with the new tax-preferred savings account provision (National Treasury, 2012). However, following public consultation, it was decided to retain the interest exemption as a transitional provision, but never to increase the monetary thresholds again, which will eventually result in the real value of the exemption being eroded by inflation15.
Examination of the salient features of the new savings incentive provision
Nature of the incentive
Tax incentives to encourage savings can either be tax-prepaid or tax-deferred incentives (Laurin & Poschmann, 2010). In tax-prepaid plans, contributions are made from post-tax earnings and no deduction is allowed for contributions made. Tax-deferred plans allow for contributions to be made from pre-tax income by providing a deduction for contributions made. Both the existing non-retirement savings incentives, that is, the tax-exempt interest threshold (section 10(1)(i)) and tax-free savings accounts (section 12T), are tax-prepaid plans. Conversely, the incentives to encourage contributions to retirement funds are tax-deferred plans (section 11(k))16, which allows for a limited deduction of
contributions to retirement funds.
13 Dividends are exempt from normal tax but subject to dividends withholding tax of 20% (section 64E of the Income Tax Act (58 of
1962)).
14 Distributions received from real estate investment trusts (REITs) are subject to normal tax (section 25BB of the Income Tax Act (58 of
1962)).
15 South Africa’s consumer price index (CPI) was 4.8% during October 2017 (Statistics South Africa, 2017).
16 Section 11(k) is expected to be replaced with the similarly-worded section 11F when the 2017 Taxation Laws Amendment Bill is signed
Contribution limit
The new savings incentive provision contained within section 12T of the Income Tax Act (58 of 1962) allows natural persons17 to make a maximum contribution of up to R33 00018 per tax year19 to
specifically designated “tax-free investments”. An aggregate (lifetime) contribution limit of R500 000 also applies (section 12T(4)). The lifetime contribution limit only applies in respect of contributions, and investment returns capitalised do not deplete this limit. The lifetime limit serves as a mechanism for government to control the cost of the incentive and will be increased only when the fiscus can afford it (National Treasury, 2014).
Normal tax, capital gains tax and dividends tax exemption
Any amount received or accrued in respect of a tax-free investment is exempt from normal tax (section 12T(2)), and investment returns do not deplete the contribution limits (section 12T(5)). Any capital gain or loss determined on disposal of an investment is disregarded for capital gains tax20
purposes (section 12T(3)). A dividend paid in respect of a tax-free investment is exempt from dividends tax (section 64F(1)(o)).
Penalty clause for excess contributions
Product providers are not permitted to accept investments in excess of the annual or lifetime contribution limits21. The provisions of section 12T and the related regulations, however, do not
prohibit holding tax-free investments with multiple product providers. For that reason, section 12T(7) imposes a severe penalty of 40% of contributions made in excess of the annual or lifetime contribution limits.
Evaluation of new savings incentive provision against design criteria
Equity
Smith (1776/1892) recorded the equity principle as follows: “[t]he subjects of every state ought to contribute towards the support of the Government, as nearly as possible, in proportion to their respective abilities, that is, in proportion to their revenue which they respectively enjoy under the
17 The incentive is also available to deceased estate or insolvent estate of a natural person that is deemed to be one and the same
person as that natural person in respect of the contributions made by that person.
18 The limit was R30 000 upon initial implementation of the statute on 1 March 2015. The increased annual contribution limit of R33 000,
hailed as an inflationary adjustment, became effective on 1 March 2017.
19 The South African tax year (“year of assessment”) for natural persons ends on the last day of February each year.
20 Forty per cent of capital gains derived by natural persons are included in their taxable income and taxed accordingly. A Natural person
is however entitled to an annual exclusion of R40 000 on capital gains derived.
protection of the State” (p. 454). The principle of equity is discussed by many from one of two perspectives, that is, horizontal equity or vertical equity (Institute on Taxation and Economic Policy, 2012; Vivian, 2006).
According to Musgrave (1990), horizontal equity calls for equal tax treatment of taxpayers in equal circumstances. Section 9 of the Constitution also preserves this principle guaranteeing everyone the right to equal benefit of the law. Institute on Taxation and Economic Policy (2012) states that, where a tax system fails the test of horizontal equity, public support for the system is undermined, the system is harder to defend and taxpayer honestly is likely to decline. The new savings incentive is equally available to all individual taxpayers and, unlike the predecessor incentive, extends wider than merely interest-bearing investments, and underlying investment assets can include shares, commodities, real estate investment trusts and, to a limited extent, derivative financial instruments.
Sneed (1965) describes vertical equity as the way taxpayers, with dissimilar abilities to pay, must be treated and is often achieved through progressive tax systems where wealthier individuals are expected to carry a greater tax burden. Because the section 12T incentive allows every individual to contribute the same monetary amount per tax year and per lifetime, the extent of the benefit derived from the incentive is largely driven by the return on the investment instead of the effective income tax rate of the specific taxpayer22. Despite the exemption from normal tax, capital gains tax and dividends
tax, the new incentive does not offer an escape from estate duty23. The Davis Tax Committee (DTC,
2015) contends that estate duty is the only wealth tax in South Africa. The wealthy therefore do not benefit excessively from the new incentive.
Effectiveness at promoting new savings by target groups
The objective of the section 12T savings incentive is to encourage new savings by households with
low to moderate levels of taxable income. The regulations24 governing tax-free investments prohibit
the reclassification or conversion of pre-existing financial instruments or policies as tax-free investments. While this does not preclude an investor from withdrawing an existing investment and re-investing the proceeds in a tax-free investment, the potential tax consequences and administrative burden associated with the withdrawal could serve to discourage this behaviour.
22 Income tax legislation previously allowed for a tax deduction for certain medical expenditure including contributions to medical aid
schemes. This system was viewed as inequitable, because it provided a greater benefit to higher income taxpayers (National Treasury, 2011). This system has since been replaced with a tax credit.
23 Estate duty is levied at a rate of 20% (Section 2(2) and the First Schedule to the Estate Duty Act (45 of 1955)) of the “dutiable amount”
of an estate and no estate duty is levied on the first R3.5 million net value of an estate (Section 4A of the Estate Duty Act (45 of 1955)).
Section 12T does not contain any provision to ensure that the incentive is taken up by low to moderate taxable income households, but as discussed earlier, the incentive does not excessively favour the wealthy. Intellidex (2016) estimate that just over a fifth of the 262 000 tax-free savings accounts opened in the first year of the incentive25 were opened by first-time savers.
Economic efficiency and cost effectiveness
“Tax expenditures” is the term given to revenue foregone as a result of legislative provisions that deviate from the standard tax treatment (National Treasury, 2017). Government does not currently measure the revenue foregone because of the section 12T incentive, but plans to do so in future (National Treasury, 2017). According to Intellidex (2016), the first year of the incentive saw R2.6 billion invested in tax-free investments. Based on the assumption that the average rate of return on investments amounted to 6.589% (refer Table 1) and that the average marginal tax rate of investors is 25%, one could estimate26 the revenue foregone at R43 million during the first year of operation of
the incentive. However, since the section 12T savings incentive is a tax-prepaid plan with limited annual contributions, the resultant tax expenditure will initially be much lower than the current interest exemption and the real associated tax expenditure will only be borne by the fiscus in future.
Revenue foregone is not the only cost to the fiscus. There is also the indirect cost of administering the incentive. The new incentive will arguably be more cumbersome to administer when compared with the interest exemption, especially activities conducted to monitor compliance with contribution limits and imposing penalties for non-compliance. According to Smith (1776/1892), “every tax ought to be contrived as both to take out and keep out of pockets of the people as little as possible over and above what it brings, into the public treasury of the State” (p. 454). The cost of revenue collection amounted to 0.96% of revenue collected in 2015/2016, which is marginally within the international benchmark of 1%.
25 Tax year ending 29 February 2016.
26 This is observably a very rough estimate, because dividends, capital gains and other income are taxed at different rates and the
estimate also disregards the fact that some of the investments could have been made only at the end of the tax year. Even tax expenditure data published by government is qualified, as it is heavily dependent on assumptions and the methodology used (National Treasury, 2017). This estimate merely serves to illustrate that the initial tax expenditure of the section 12T incentive is significantly lower than on the interest exemption.
Table 1: Interest rates on cash tax-free investments
Product provider Nominal interest rate Average rate of return
Standard Bank 6.15% to 7.89% 7.070%
Absa Bank 4.50% to 7.50% 6.000%
Capitec 5.35% to 7.85% 6.600%
FNB 6.15% to 7.40% 6.775%
Nedbank 6.25% to 6.75% 6.500%
Average rate of return 6.589%
Certainty, transparency, convenience, simplicity and suitability
The tax which a person has to pay should be certain and not arbitrary and should be imposed at the time and in the manner most convenient to the taxpayer (Smith, 1776/1892). The provisions of the savings incentive appear to be clear and do not appear to contain significant uncertainties.
A minor uncertainty is pointed out by De Koker and Williams (2017) who noted that it is uncertain whether trustees of a trust may invest capital on behalf of a beneficiary in a tax-free investment to which such a beneficiary has a vested right. Since the definition of a “tax-free investment” in section 12T(1) refers to a natural person as the owner as opposed to beneficial owner, such an investment will likely not fall within the ambit of section 12T.
National Treasury (2012) initially proposed a transitional dispensation allowing taxpayers aged 45 and older to invest a greater portion of their lifetime limit during the first two years of the incentive, which would have made the new incentive more convenient for older taxpayers. However, the decision to retain the interest exemption negated the need for a transitional dispensation.
The transparency principle requires costs and fees related to the investment to be clear from inception of the product. The regulations27 that govern tax-free investments require product providers
to disclose fees, charges and costs payable to an investor.
The principle of simplicity requires the investment product to be simple to understand in terms of identifying the investment assets and returns, and the principle of suitability requires the product to be suitable for the “average person” and not be of a speculative nature. The regulations28 governing
tax-free investments contain certain limitations with respect to the composition of investments. No more than 10% of the value of a tax-free investment may be derived from a single company, 80% of
27 Regulation 6(2)(b). 28 Regulations 13 and 14.
any shares that make up the assets of a tax-free investment must be listed, and derivative financial instruments may only be used for the reduction of risk of loss or cost.
Visibility
Visibility or awareness of tax-free investments is an important attribute or criteria to ensure that the incentive is well targeted. For this reason, product providers must29 specifically advertise and
designate a tax-free investment as being “tax-free”. Accordingly, many financial institutions offer and advertise tax-free investments, including the most prominent South African banks (refer Table 1).
Prevention of tax arbitrage
In order to obtain a deduction for the incurral of interest, section 24J(2) requires a taxpayer to incur such interest in the production of “income”. Income is defined in section 1(1) as “the amount remaining of the gross income of any person for any year or period of assessment after deducting therefrom any amounts exempt from normal tax” (p. 17). Since the income derived from a tax-free investment is exempt from normal tax, a taxpayer incurring interest to fund the acquisition of a tax-free investment would be precluded from deducting such interest. Furthermore, section 23(f) prohibits the deduction of any expenses incurred in respect of amounts received or accrued that do not constitute “income”. It is therefore submitted that extant South African income tax legislation contains sufficient mechanisms to prevent tax arbitrage in respect of tax-free investments.
2.7 COMPARISON TO THE INCENTIVE AVAILABLE IN CANADA
Characteristics of the Canadian incentive
Tax-free savings accounts were introduced in Canada in January 2009 (Alarie, 2009) in order to provide an incentive for Canadians to save more (Department of Finance Canada, 2012). These TFSAs do not provide for a tax deduction for the contributions made to the savings account but investment returns: interest, dividends and capital gains are tax-free (Kesselman, 2015). Contribution and investment earnings are exempt from tax on withdrawal, and funds can be withdrawn at any time (Kokorian, 2008). When a TFSA holder dies, there might be tax implications depending on certain factors such as the type of beneficiaries (Department of Finance Canada, 2012).
The definition of a “successor holder”, in provinces that allow the beneficiary designation, is the spouse or common-law partner of the original TFSA holder (Department of Finance Canada, 2012).
If a beneficiary has been designated as a successor holder there will not be any tax implications on the death of the original TFSA holder (Department of Finance Canada, 2012). The successor holder, named in the will of a person, will after the death of the TFSA holder become the new holder (Department of Finance Canada, 2012). He is then allowed to transfer the amount to his existing TFSA without affecting his contribution room (Department of Finance Canada, 2012).
A resident of Canada who is 18 years and older may invest in a TFSA (Alarie, 2009). Initially, contributions had been limited to $5 000 per tax year to be adjusted with inflation, but the limit was increased with inflation in 2013 to $5 500 (Department of Finance Canada, 2012), to $10 000 in 2015, but back again to $5 500 in 2016 (Kesselman, 2015). The limit of the contribution indicates that the purpose is for short-term savings and not savings for retirement (Alarie, 2009).
Excess contributions (above the limit) are subject to tax at 1% per month (Kokorian, 2008). Unused contributions can be carried forward indefinitely (Laurin & Poschmann, 2010). The limit remains the same for all natural persons irrespective of their income (Kesselman, 2015). Any amount withdrawn can also be put back later, and individuals who make withdrawals may recontribute withdrawals in future years (Kokorian, 2008). The fact that withdrawals can be recontributed indicates the use of TFSAs for short-term savings and consumption without losing cumulative contribution limits (Kesselman, 2015).
Qualifying investments include mutual funds, publicly-traded securities, government and corporate bonds, etc. (Kokorian, 2008). The type of accounts that can be invested in may not be very risky with very high returns to ensure that taxpayers use the account for savings and not for realising high returns on investments tax-free (Alarie, 2009).
According to Scarlat and Lefebvre (2008), the TFSA has the potential to increase savings for all income groups (high and low earners) and the purpose is for savings other than savings for retirement.
Contributions can come from the person self or from someone else (Kesselman, 2015). A wealthy person can therefore make a donation to a relative, which can be invested in a TSFA (Alarie, 2009) or contributed to a TFSA on their behalf (Kesselman, 2015) with no tax consequences. If a donation like this substitutes a transfer of wealth that would have been made on death, and where tax would have been paid, it would result in the transfer of wealth without paying any tax on this (Alarie, 2009).
Comparison of the South African tax incentive to that of Canada
The TFSA in Canada is similar to the tax-free investment available in South Africa in that it does not provide for a deduction for contributions made to the savings account but investment returns: interest, dividends and capital gains are also tax-free and withdrawals can be made tax-free at any time. Another similarity is that a person who invests in this savings account should be a resident. However, there is no requirement for a person to be 18 years old to be able to invest in a tax-free investment in South Africa; therefore, any person may invest in a tax-free investment irrespective of their age. For both the TFSA in Canada and the tax-free investment in South Africa, there are set contribution limits that apply, and contributions made in excess of the contribution limits are subject to penalties. Taxpayers are, however, allowed to contribute to more than one account, but subject to the contribution limits. Similar to the TFSA in Canada, a tax-free investment in South Africa will be subject to estate duty on the death of a person, except if the investment is transferred to a taxpayer’s spouse. The TFSA in Canada differs from the tax-free investment in South Africa in that individuals are allowed to carry forward any unused contributions, which is not allowed in South Africa. Unused contributions in a year may not be carried forward to the next tax year. In Canada withdrawals may be recontributed in future years, where contributions made to a tax-free investment in South Africa are subject to the R33 000 annual limit. The TFSA in Canada also does not have the lifetime limit, but would not be necessary, as withdrawals may be replaced in future years.
In conclusion, there are many similarities between the TFSA in Canada and the tax-free investment in South Africa. Recommendations which could add value to the tax-free investment in South Africa would be to allow for unused contributions to be carried forward indefinitely. If a taxpayer therefore does not contribute at the full limit per year, for example, only make a contribution of R13 000, the taxpayer will be allowed to contribute R53 000 (R20 000 prior year + R33 000 current year) in the next tax year. This will give a person the opportunity to increase his savings in future years. If this is allowed, however, a starting age would have to be implemented similar to the 18-year requirement for TFSAs in Canada. The implication of this would be that parents would no longer be allowed to invest in a tax-free investment in South Africa on behalf of their children.
Withdrawals could also be allowed to be replaced in future years. The fact that withdrawals may be recontributed indicates the use of TFSA in Canada for short-term savings and consumption without losing cumulative contribution limits, which adds to the fact that the tax-free investment in South Africa will only be beneficial in comparison with other savings vehicles if it is used as a long-term investment. Currently a person may be hesitant to invest in a tax-free investment because if the
amounts are required and withdrawn, the person’s lifetime limit will be reduced, which again may decrease savings to begin with. If withdrawals could be replaced in future years, a person would not need to be concerned about his lifetime limit decreasing which, in turn, would serve as motivation towards savings.
2.8 CONCLUSION AND RECOMMENDATIONS FOR FURTHER RESEARCH
To combat the decline in household savings prevalent in South Africa over the past 50 years, the South African government has taken steps to ameliorate this problem by offering an incentive through the income tax system by way of tax-preferred savings accounts. An immediate concern is that the South African income tax base is exceedingly narrow with only 8.5%30 of the population subject to
income tax, resulting in a restricted target audience.
Extant theory on the use of tax incentives to drive non-tax behaviour indicates that such incentives often result in a distortion of investment decisions and are ineffective and expensive. This study identified attributes or criteria to serve as benchmarks for evaluating the new tax incentive to promote household savings. The attributes identified are that the incentive must promote new savings among low- to moderate-income earners, it must be cost-effective, it must be visible, product offerings must be simple, transparent and suitable, and it must prevent tax arbitrage. In addition, it must conform to the principles of equity, certainty, convenience, and economic efficiency.
The evaluation found that, although the incentive has some mechanisms in place to encourage new savings, only a fifth of accounts opened during the first year of operation are estimated to be from new savers. While the incentive does not disparagingly benefit wealthy individuals, it does not contain any features to ensure use by low to moderate taxable income households (the target group). The incentive has both a direct cost (revenue foregone) and an indirect cost (increased administration cost) to the fiscus. Government does not yet measure the tax expenditure on this incentive, but plans to do so in future. This study estimated that, owing to the design of the incentive where contributions are limited, the initial cost of the incentive to the fiscus will be significantly lower than the current interest-free exemption, and the real cost of the new incentive will only be borne by the fiscus in the future.
This study found that the new incentive is designed to be visible, simple, transparent and suitable. It is more equitable than its predecessor, because it allows for a wider range of investments to be made
30 South Africa assessed 4 788 334 individual taxpayers in the 2015 tax year (South African Revenue Service, 2016) while the 2016
tax-free. The incentive does not infringe on the certainty and convenience principles, but it remains unclear whether the incentive will be economically efficient and cost-effective.
An evaluation of the incentive against an established comparable Canadian incentive revealed that it could be beneficial to allow for unused contributions to be carried forward, as this might lead to increased future savings. In addition, allowing replacement of withdrawals will likely enhance short-term savings, as the current incentive is more suitable for long-short-term investment.
REFERENCES
Alarie, B. (2009). Policy forum: Assessing tax-free savings accounts – Promises and pressures.
Canadian Tax Journal, 2009(3), 504-532.
Alley, C., & Bentley, D. (2005). A remodelling of Adam Smith’s tax design principles. Retrieved from http://epublications.bond.edu.au/cgi/viewcontent.cgi?article=1044&context=law_pubs
Antolín, P., De Serres, A., & De la Maisonneuve, C. (2004). Long-term budgetary implications of tax-favoured retirement saving plans. OECD Economic Studies, 2(39), 25-72.
Aron, J., & Muellbauer, J. J. (2000). Personal and corporate saving in South Africa. The World
Bank Economic Review, 14(3), 509-544.
Attanasio, O. P., Banks, J., & Wakefield, M. (2004). Effectiveness of tax incentives to boost
(retirement) saving: Theoretical motivation and empirical evidence. OEDC Economic Studies,
2004(2), 146-172.
Blough, R. (1955). The history and philosophy of taxation. Williamsburg: College of William and Mary.
Bolnick, B. (2004). Effectiveness and economic impact of tax incentives in the SADC region, USAID, Technical Report.
Brunori, D. (1997). Principles of tax policy and targeted incentives. State and Local Government
Review, 29(1), 50-61.
Constitution see South Africa (1996).
Cronje, R. (2017). Gigaba faces R50 billion budget shortfall – economist. Fin24. Retrieved from https://www.fin24.com/Economy/gigaba-faces-r50-billion-budget-shortfall-economist-20170901
Davis Tax Committee (DTC) see South Africa (2015).
De Koker, A.P., & Williams, R.C. (2017). Silke on South African Income Tax. Lexisnexis.
Department of Finance Canada. (2012). Canada, Tax Expenditures and Evaluations 2012 (Ottawa: Department of Finance, 2013), 31-45.
Easson, A., & Zolt, E. M. (2002). Tax incentives. Retrieved from
http://www.siteresources.worldbank.org/INTTPA/Resources/EassonZoltPaper.pdf Employment Tax Incentive Act 26 of 2013 see South Africa (2013).
Estate Duty Act 45 of 1955 see South Africa (1955).
Finmark Trust. (2014). Results from FinScope Consumer Survey South Africa 2014, November 2014. Retrieved from http://www.finmark.org.za/wp-content/uploads/pubs/MedRel_FS_SA_ 20141.pdf
Furman, J. (2008). The Concept of Neutrality in Tax Policy. Retrieved from
Galper, H., & Steuerle, E. (1983). Tax incentives for saving. Washington D.C.: Brookings Institution. Hage, J. (2011). The method of a truly normative legal science. In M. van Hoecke (Ed.),
Methodologies of legal research (pp. 19-44). Oxford: Hart Publishing.
Income Tax Act 58 of 1962 see South Africa (1962).
Institute on Taxation and Economic Policy. (2012). Tax principles: building blocks of a sound tax system. Retrieved from http://www.itepnet.org/pdf/pb9princ.pdf
Intellidex (2016). A study of tax-free savings account takeup in South Africa. Retrieved from http://www.intellidex.co.za/wp-content/uploads/2016/06/Intellidex-TFSA-Survey-Report-Jun2016-final.pdf
Kesselman, J. R. (2015) Tax-free savings accounts: Expanding, restricting, or refining? Canadian
Tax Journal, 63(4), 905-945.
Kokorian, C. (2008). New tax-free savings account versus the old, reliable RRSP. CMA
Management, 82(7), 43-44.
Laurin, A., & Poschmann, F. (2010). Saver’s choice: Comparing the marginal effective tax burdens on RRSPs and TFSAs. Fiscal and Tax Competitiveness/Pension Papers. Retrieved from https://ssrn.com/abstract=1581672.
Lynn, A. D. (1976). Adam Smith’s fiscal ideas: An eclectic revisited. National Tax Journal, 29(4), 369-378.
McKerchar, M. (2009). Philosophical paradigms, inquiry strategies and knowledge claims: Applying the principles of research design and conduct to taxation. University of New South Wales
Faculty of Law Research Series, 31(1), 3-22.
Musgrave, R. A. (1990). Horizontal equity once more. National Tax Journal, 43(2), 113-122. National Treasury see South Africa.
OED (2017). Household savings (indicator). Retrieved from https://data.oecd.org/hha/household-savings.htm
Prinsloo, J.W. (2000). The saving behaviour of the South African economy. South African Reserve Bank. Retrieved from
https://www.resbank.co.za/Lists/News%20and%20Publications/Attachments/2513/occ14fin.p df
Phungo, R. (2015). University fees: Free higher education is possible in South Africa. Retrieved from http://irr.org.za/reports-and-publications/articles-authored-by-the-institute/university-fees-free-higher-education-is-possible-in-south-africa-2013-biznews-21-october-2015
Reese, T. (1980). The politics of taxation. Westport: Quorum.
Scarlat, M., & Lefebvre, R. (2008). Tax-free savings accounts – Shifting opportunity. Canada: Certified General Accountants Association of Canada.
Smith, A. (1892). An inquiry into the nature and causes of the wealth of nations. London: Routledge. (Original work published 1776)
Smits, J. M. (2012). The mind and method of the legal academic. Cheltenham: Edward Elgar Publishing.
Sneed, J. T. (1965). The criteria of federal income tax policy. Stanford Law Review, 17(4), 567-613. South Africa. (1955). Estate Duty Act 45 of 1955. Pretoria: Government Printer.
South Africa. (1962). Income Tax Act 58 of 1962. Pretoria: Government Printer. South Africa. (1991). Value-added Tax Act 89 of 1991. Pretoria: Government Printer.
South Africa. (1996). Constitution of the Republic of South Africa 1996. Pretoria: Government Printer.
South Africa. (2013). Employment Tax Incentive Act 26 of 2013. Pretoria: Government Printer. South Africa. Davis Tax Committee (DTC). (2015). First Interim Report on Estate Duty for the
Minister of Finance. Pretoria: Government Printer.
South Africa. Department of National Treasury. (2011). Conversion of medical deductions to
medical tax credits. Pretoria: Government Printer.
South Africa. Department of National Treasury. (2012). Incentivising non-retirement savings. Pretoria: Government Printer.
South Africa. Department of National Treasury. (2014). Non-retirement savings: Tax-free savings
accounts. Pretoria: Government Printer.
South Africa. Department of National Treasury. (2015). Notice on Regulations in terms of section 12T(8) of the Income Tax Act, 1962, on the requirements for tax free investment. (Notice 172 of 2015). Government Gazette, 38509, pp. 3-16.
South Africa. Department of National Treasury. (2017). Budget Review 2017. Pretoria: Government Printer.
South African Revenue Service. (2016). Tax Statistics for 2016. Retrieved from
http://www.sars.gov.za/Media/MediaReleases/Pages/29-November-2016---Tax-Statistics-for-2016.aspx
Statistics South Africa. (2016). Mid-year population estimates 2016. Retrieved from https://www.statssa.gov.za/publications/P0302/P03022016.pdf
Statistics South Africa. (2017). Key findings: P0141 – Consumer Price Index (CPI), October 2017. Retrieved from http://www.statssa.gov.za/?page_id=1856&PPN=P0141&SCH=6766 Trading Economics. (2017). South Africa GDP Growth Rate. Retrieved from
https://tradingeconomics.com/south-africa/gdp-growth
UN Statistics Division. (n.d.). Methodology. Standard country or area codes for statistical use
(M49). Retrieved from https://unstats.un.org/unsd/methodology/m49/
Van Hoecke, M. (2011). Legal doctrine: which method(s) for what kind of discipline? In Van Hoecke, M. (Ed.), Methodologies of legal research. Oxford: Hart Publishing.
Vivian, R. W. (2006). Equality and personal income tax – The classical economists and the Katz Commission. South African Journal of Economics, 74(1), 79-109.
World Bank. (2011). South Africa Economic Update. Focus on Savings, Investment, and Inclusive
Growth, July 2011. Retrieved from
http://www-wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2011/08/01/000386194_20 110801012714/Rendered/PDF/635390WP0SAEU000BOX361521B00PUBLIC0.pdf
CHAPTER 3: CONCLUSION
3.1 OBJECTIVE OF THIS CHAPTER
The objective of this chapter is to provide a summary of the key findings of the study and indicate how these findings address the research question formulated in Chapter 1.
3.2 RESEARCH FINDINGS
3.2.1 Research objective 1
The first research objective was to determine appropriate design attributes or criteria to serve as benchmarks for the evaluation of an effective savings tax incentive.
The attributes identified are that the incentive must promote new savings among low- to moderate-income earners, it must be cost-effective, it must be visible, product offerings must be simple, transparent and suitable, and it must prevent tax arbitrage. In addition, the new tax incentive must conform to the principles of equity, certainty, convenience, and economic efficiency.
3.2.2 Research objective 2
The second research objective was to evaluate the design of section 12T against the attributes identified to determine whether the design of the statute is fundamentally effective to promote household savings in South Africa.
The evaluation found that, although the incentive has some mechanisms in place to encourage new savings, only a fifth of accounts opened during the first year of operation are estimated to be from new savers. While the incentive does not disparagingly benefit wealthy individuals, it does not contain any features to ensure use by low to moderate taxable income households (the target group). The incentive has both a direct cost (revenue foregone) and an indirect cost (increased administration cost) to the fiscus. Government does not yet measure the tax expenditure on this incentive, but plans to do so in future. This study estimated that, owing to the design of the incentive where contributions are limited, the initial cost of the incentive to the fiscus will be significantly lower than the current interest-free exemption, and the real cost of the new incentive will only be borne by the fiscus in the future.
The study found that the new incentive is designed to be visible, simple, transparent and suitable. It is more equitable than its predecessor because it allows for a wider range of investments to be made
